Kernel panic right after routine update/upgrade on June 20, 2023
I had just installed shotcut in order to learn how to edit some MP4 movie clips
that I just made. After that step I ran apt-get update and apt-get upgrade, whereupon
Software Updater chimed in, so I ran that and the following standard cleanup.
Upon rebooting, a kernel panic intervened. Next, I chose GRUB's 2nd menu choice,
but not one of the "repair" choices. This second-choice bootup went without incident.
What's next in the usual order of steps to correct whatever went wrong with the
update/upgrade process ? I'm running Trisquel 11 on a Thinkpad T420.
The Shotcut install still seems to be present, but I haven't tried it.
After that step I ran apt-get update and apt-get upgrade, whereupon Software Updater chimed in, so I ran that and the following standard cleanup.
A newer image of Linux was installed, wasn't it? Read the end of /var/log/apt/history.log to know.
Upon rebooting, a kernel panic intervened.
Is it reproducible?
Next, I chose GRUB's 2nd menu choice, but not one of the "repair" choices.
As far as I understand, you booted a previous version of the kernel. Do not remove it before being certain that a newer version works!
Here's what the ending paragraphs of history.log say:
Start-Date: 2023-06-20 15:27:53
Commandline: aptdaemon role='role-commit-packages' sender=':1.559'
Install: linux-headers-5.15.0-75-generic:amd64 (5.15.0-75.82+11.0trisquel23, automatic), linux-modules-5.15.0-75-generic:amd64 (5.15.0-75.82+11.0trisquel23, automatic), linux-modules-extra-5.15.0-75-generic:amd64 (5.15.0-75.82+11.0trisquel23, automatic), linux-image-5.15.0-75-generic:amd64 (5.15.0-75.82+11.0trisquel23, automatic), linux-headers-5.15.0-75:amd64 (5.15.0-75.82+11.0trisquel23, automatic)
Upgrade: linux-headers-generic:amd64 (5.15.0.73.71+11.0trisquel6, 5.15.0.75.73+11.0trisquel6), linux-generic:amd64 (5.15.0.73.71+11.0trisquel6, 5.15.0.75.73+11.0trisquel6), linux-image-generic:amd64 (5.15.0.73.71+11.0trisquel6, 5.15.0.75.73+11.0trisquel6)
End-Date: 2023-06-20 15:28:53
Start-Date: 2023-06-20 15:29:05
Commandline: aptdaemon role='role-commit-packages' sender=':1.559'
Remove: linux-modules-extra-5.15.0-72-generic:amd64 (5.15.0-72.79+11.0trisquel22), linux-image-5.15.0-72-generic:amd64 (5.15.0-72.79+11.0trisquel22), linux-headers-5.15.0-72:amd64 (5.15.0-72.79+11.0trisquel22), linux-headers-5.15.0-72-generic:amd64 (5.15.0-72.79+11.0trisquel22), linux-modules-5.15.0-72-generic:amd64 (5.15.0-72.79+11.0trisquel22)
End-Date: 2023-06-20 15:29:41
The first paragraph has the new packages; the 2nd has the cleanup.
A newer image of Linux, version 5.15.0-75, was indeed installed. Is the problem reproducible?
Yes; three times so far. On restart, the GRUB menu appears like an oasis in the desert.
Here are "Screengrabs" of the results:
The smaller one seems to pinpoint an unexpected "C" ...
The larger one is the entire page (via digital camera).
The error is exactly that of https://askubuntu.com/questions/1216141/i-am-facing-issues-with-kernel-panic-not-syncing-attempted-to-kill-init whose solution involves regenerating the initramfs image with update-initramfs. Try it.
Here are the successful details ==>
You must specify at least one of -c, -u, or -d.
Usage: update-initramfs {-c|-d|-u} [-k version] [-v] [-b directory]
Options:
-k version Specify kernel version or 'all'
-c Create a new initramfs
-u Update an existing initramfs
-d Remove an existing initramfs
-b directory Set alternate boot directory
-v Be verbose
See update-initramfs(8) for further details.
The above reference came up with ==>
Update #3:
sudo update-initramfs -c -k 5.3.0-40-generic solved the problem
Restart; let the kernel panic occur again; kill & allow restart; copy 1st non-recovery Grub selection ==>
Linux-Libre 5.15.0-75-generic
sudo update-initramfs -c -k 5.15.0-75-generic (vs. Linux-Libre 5.15.0-73-generic, which works OK)
sudo update-initramfs -c -k 5.15.0-75-generic
[sudo] password for george:
update-initramfs: Generating /boot/initrd.img-5.15.0-75-generic
I: The initramfs will attempt to resume from /dev/sdc3
I: (UUID=6dc6d6da-161d-43fe-a98b-711e7368ad1c)
I: Set the RESUME variable to override this.
george@george-ThinkPad-T420:~$ Trisquel now running Linux-Libre 5.15.0-75-generic.
Thanks to Magic Banana & the folks who redesigned grub to restart with the previous working trisquel kernel.
Oops; happened again on September 17, 2023:
A Google search on the first line of the kernel panic output:
/init: conf.d/zz-resume-auto: line 1 syntax error: unexpected "c" returned this link:
https://askubuntu.com/questions/902608/kernel-panic-after-updating-ubuntu-says-syntax-error where it's said ==>
"Perhaps it's a corrupt init ramdisk image. I managed to solve something similar this way.
First boot into a recovery environment where you have access to a shell and necessary volumes of the original system
(notably rootfs and /boot). Mount them and chroot into the original rootfs.
Delete all initramdisks:
update-initramfs -d -k all
Regenerate them:
update-initramfs -c -k all
Now reboot, problem solved."
I did that this way:
sudo update-initramfs -d -k all
update-initramfs: Deleting /boot/initrd.img-5.15.0-83-generic
update-initramfs: Deleting /boot/initrd.img-5.15.0-82-generic
george@george-ThinkPad-T420:~$ sudo update-initramfs -c -k all
update-initramfs: Generating /boot/initrd.img-5.15.0-82-generic
I: The initramfs will attempt to resume from /dev/sdc3
I: (UUID=6dc6d6da-161d-43fe-a98b-711e7368ad1c)
I: Set the RESUME variable to override this.
update-initramfs: Generating /boot/initrd.img-5.15.0-83-generic
I: The initramfs will attempt to resume from /dev/sdc3
I: (UUID=6dc6d6da-161d-43fe-a98b-711e7368ad1c)
I: Set the RESUME variable to override this.
Rebooting brought up the usual login screen.